There are no major studies on the development of Conduct Disorder (CD) in girls. As a consequence, little is known about the developmental pathways to CD, its comorbid conditions, and risk and protective factors that might explain its development with or without comorbid conditions. This proposal is for Phase 2 of a five-year follow-up of a major study on the development of Conduct Disorder in girls (N=2,451). In Phase 1 of the study, a random sample of 5-8 year-old girls was drawn from the city of Pittsburgh though a enumeration of 103,238 households. The resulting random sample for yearly follow-up consisted of four cohorts of 5, 6, 7, and 8 year-old girls (about N=600 in each of the four cohorts), with a slight overrepresentation of girls in the most disadvantaged neighborhoods to increase the eventual yield of CD in the study. So far, the girls, their parent (usually the mother), and their teacher have been assessed twice at a yearly interval, and a third assessment is planned for the third and final year of Phase 1. Participant cooperation in the second assessment which is not yet completed has been very high (over 95%). In the proposed Phase 2 of the study, when the girls will be between the ages of 8 and 15, we plan to continue yearly assessments. We anticipate that during this period, the prevalence of CD will increase, as will the prevalence and severity of other disruptive behaviors including delinquency, internalizing problems, particularly depressed mood and PTSD, and symptoms of somatoform disorder. The aims of Phase 2 of the study are: 1) Examine the content, convergent and predictive validity of conduct problems and diagnosis of Conduct Disorder in girls; 2) Identify developmental models of Conduct Disorder and disruptive behavior; 3) Identify the role of comorbid problems in the development and course of conduct problems; and 4) Produce a model of risk and protective factors that is developmentally specific and that can explain why many problems in girls emerge and worsen after puberty.